The use of robotic systems in automated manufacturing has become increasingly important in high technology device manufacturing. Robots can often perform repeated tasks with precision and efficiency that is generally not achievable through the use of human labor. Further, robots can be used in environments where manufacturing techniques for picking, placing, and assembling very small components make the use of human labor undesirable.
For example, this is particularly important in disk drive manufacturing where misplaced components may result in scraping during the assembly process possibly contaminating the disk drive and resulting in disk drive failure. This is particularly true for small-form-factor disk drives that require very high accuracy assembly techniques.
Robots used in disk drive manufacturing by picking, placing, and assembling components are generally well known in the art. However, in present robotic disk drive manufacturing, components are picked and placed by a robot from a taught point using a gripper, which may be prone to error.
In order to ensure precise positioning of components moved by the robot, reference points or coordinates for a desired or predetermined position of the robot's end effector, such as a gripper, may be entered into a memory of a robot controller as part of a calibration procedure. Acquiring the reference coordinates generally involves jogging the gripper to predetermined positions through a manual procedure. Arrival of the robot's end effector, such as a gripper, at the predetermined position may be confirmed by manually observing the end effector position or by having the end effector trigger a sensor.
Typically the movement of the robot's end effector for calibration is done manually. An operator must observe the location of the end effector relative to an object or target within the processing system to estimate the position of the end effector.
Thus, although attempts to improve robot accuracy are presently being made, these present techniques are typically tedious, time-consuming, undesirably expensive, involve human intervention, and often do not provide an acceptable level of accuracy. Unfortunately, for high technology device manufacturing, such as disk drive manufacturing, that requires very high accuracy to avoid scraping and contamination, these present techniques are not acceptable.
A need therefore exists for improved calibration techniques for vision based robotic systems.